US8212939B2 - Non-intrusive determination of an objective mean opinion score of a video sequence - Google Patents

Non-intrusive determination of an objective mean opinion score of a video sequence Download PDF

Info

Publication number
US8212939B2
US8212939B2 US12/094,084 US9408408A US8212939B2 US 8212939 B2 US8212939 B2 US 8212939B2 US 9408408 A US9408408 A US 9408408A US 8212939 B2 US8212939 B2 US 8212939B2
Authority
US
United States
Prior art keywords
interframe
calculating
similarities
sequence
pausing
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related, expires
Application number
US12/094,084
Other languages
English (en)
Other versions
US20080284853A1 (en
Inventor
Pero Juric
René Widmer
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Swissqual License AG
Original Assignee
Swissqual License AG
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Swissqual License AG filed Critical Swissqual License AG
Assigned to SWISSQUAL LICENSE AG reassignment SWISSQUAL LICENSE AG ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: WIDMER, RENE, JURIC, PERO
Publication of US20080284853A1 publication Critical patent/US20080284853A1/en
Assigned to SWISSQUAL LICENSE AG reassignment SWISSQUAL LICENSE AG CHANGE OF ASSIGNEE'S ADDRESS Assignors: SWISSQUAL LICENSE AG
Application granted granted Critical
Publication of US8212939B2 publication Critical patent/US8212939B2/en
Expired - Fee Related legal-status Critical Current
Adjusted expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N17/00—Diagnosis, testing or measuring for television systems or their details
    • H04N17/004—Diagnosis, testing or measuring for television systems or their details for digital television systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/142—Detection of scene cut or scene change
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/154—Measured or subjectively estimated visual quality after decoding, e.g. measurement of distortion
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
    • H04N19/61—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding in combination with predictive coding

Definitions

  • the invention relates to a method for assessing the quality of a video sequence.
  • the method automatically attributes an “objective mean opinion score” MOSo to the sequence, which is a score value that is indicative of a mean opinion score MOS that the sequence is expected to have when viewed by a group of human watchers.
  • the assessment of the quality of a video sequence is of importance when characterizing the performance of a video distribution network, of a transmission or compression algorithm, or of any other hard- or software involved in the creation, transmission, storage, rendering or handling of video data.
  • the original video sequence may or may not be available when assessing the quality of a given sequence.
  • the present invention relates to so called “no reference models”, where the quality of a potentially imperfect video sequence has to be derived without knowledge of the original video sequence.
  • an unknown video sequence is captured at the receiving end of a video transmission channel and then used as an input for video quality rating.
  • the main output is a video rating score or objective video MOS.
  • the video quality rating is carried out by a group of human watchers, where each member of the group attributes an opinion score to the sequence.
  • the scores of a plurality of test watchers can be averaged to obtain a mean opinion score.
  • interframe similarities s(i) are calculated for the frames of the video sequence, each interframe similarity s(i) being indicative of the similarity of two consecutive frames i ⁇ 1, i.
  • an analysis of these interframe similarities allows the determination of various parameters that are characteristic for the mean opinion score MOS.
  • the similarities s(i) can be used in the calculation of the MOSo.
  • interframe similarities s(i) is the determination of the downsampling factor DS, which in turn can be used for calculating said score MOSo.
  • Another advantageous application of the interframe similarities s(i) is the determination of the pausing parameter p, which in turn can also be used for calculating said score MOSo.
  • the pausing parameter is indicative of any pausing in the sequence.
  • a further advantageous application of the interframe similarities s(i) is the determination of a keyframe distance parameter KFD, which in turn can also be used for calculating said score MOSo.
  • This parameter is of importance for video sequences where some of the frames of the sequence were transmitted as keyframes and some of the frames were transmitted as non-keyframes, wherein the keyframes carry full information for creating a frame without reference to a prior frame and the non-keyframes carry incremental information for creating a frame from a previous frame.
  • the keyframes can be identified by checking if the interframe similarity s(i) for a frame lies in a given range. From the interframe similarities s(i) at the keyframes a keyframe distance parameter KFD can be determined, which e.g. describes how much, on an average, the keyframes differ from their previous frames. A large difference indicates a poor video quality.
  • the invention also relates to the use of this method for determining the quality of a video transmission channel.
  • FIG. 1 is a block diagram of an embodiment of a method incorporating the present invention
  • FIG. 2 is a histogram showing the result of binning the H, S and V values of a frame
  • FIG. 3 shows a typical interframe similarity s(i) for a video sequence
  • FIG. 4 illustrates the transmission of a video sequence with downsampling
  • FIG. 5 is a histogram of the binned interframe distances for a downsampling factor of 1,
  • FIG. 6 is a histogram of the binned interframe distances for a downsampling factor of 2,
  • FIG. 7 is a histogram of the binned interframe distances for a downsampling factor of 1.33
  • FIG. 8 is a histogram of the binned interframe distances for a downsampling factor of 2.4
  • FIG. 9 is a histogram of the binned interframe distances in the presence of pausing
  • FIG. 10 shows the second derivative of the interframe similarity of a higher quality sequence
  • FIG. 11 shows the second derivative of the interframe similarity of a lower quality sequence
  • FIG. 12 shows the relation between the downsampling factor DS and the perceived video quality MOS.
  • Downsampling is the rate by which a video sequence is downsampled during transmission, a measure that is used intentionally by many transmission techniques to reduce transmission bandwidth. Downsampling involves a skipping of individual frames, wherein the skipped frames are distributed substantially regularly over time. For example, each second frame may be skipped, which corresponds to a downsampling factor of 2.
  • “Pausing” is, in some way, similar to downsampling in that one or more frames are not received properly and can therefore not be displayed. However, in contrast to downsampling, pausing is an unintentional effect and skipped frames are not distributed regularly.
  • the method described here contains various parts, which all contribute to the objective mean opinion score MOSo:
  • FIG. 1 illustrates these parts.
  • the input for the method is a video file containing an uncompressed video sequence consisting of a succession of frames. Each frame is e.g. encoded in the RGB or YUV format. A typical length of the video sequence is 6 to 15 seconds.
  • the input video sequence is first passed through a preprocessing step, which transforms the frames into other formats, such as YCbCr, HSV and RGB, each of which has certain advantages depending on the analysis that follows.
  • the result of the preprocessing step is fed to a first block called temporal analysis and a second block called spatial analysis.
  • the primary purpose of the temporal analysis is the determination of the interframe similarities of the individual video frames and of parameters derived from the same. In particular, it calculates the keyframe distance, the frame rate downsampling factor and video pausing.
  • the primary purpose of the spatial analysis is the determination of spatial parameters describing each frame individually.
  • the spatial parameters can e.g. describe the average blurriness and noise level of the frames.
  • Various suitable techniques that can be used for spatial analysis are known to the person skilled in the art. The details of the spatial analysis are of no importance to the present invention.
  • Typical video quality parameters measured in a spatial domain are blurring and blockiness (blocking). Examples for Blocking and blurring detection are described in the following papers:
  • a further step titled perceptual mapping combines the parameters calculated from the temporal and spatial analysis to a single video quality number called objective mean opinion score MOSo. Further details of the perceptual mapping are described below.
  • preprocessing step As mentioned above, a main purpose of the preprocessing step is the conversion of the input frame format to HSV, YCbCr and RGB. Typical formats for uncompressed video signals are RGB and YUV. Additionally, preprocessing involves some video preparation in order to avoid the processing of the video frames containing unique color. Additionally a grade of similarity between two consecutive video frames, the interframe similarity, is calculated as a part of preprocessing step. Its result is then used in the temporal analysis step as described later.
  • the video sequence is a series of frames i, which generally differ from each other (unless two frames are identical, e.g. due to pausing or downsampling).
  • an interframe similarity s(i) is attributed to each frame i. It describes how much frame i differs from previous frame i ⁇ 1.
  • interframe similarity results can also be used in the spatial analysis acting as an indicator whether results from previous frames just can be repeated or an analysis of the new frame is required.
  • interframe similarities s(i) can be used. They differ in robustness, computational effort and sensitivity to certain types of frame changes. In the following, we describe two of them, one of them using the Y-component of the YCbCr representation of the frames, the other analyzing a histogram of the HSV representation of the frames.
  • One of the simplest methods for comparing two frames is to calculate an average value Y i-1 of all pixels belonging to a frame i ⁇ 1 using only the Y component of YCbCr signal and do the same for the next frame i. Then, the normalized difference ⁇ , which can be used as the interframe similarity s(i), can be calculated from
  • TOt H and Tot V designate the horizontal and vertical extension of each frame in pixels
  • Y h,v,i is the Y-value of the pixel at horizontal offset h and vertical offset v in frame i.
  • the interframe similarity as defined in Eqs. (1) to (3) is 0 if two consecutive frames i ⁇ 1 and i are equal and ⁇ 0 if not.
  • a more advanced method for calculating the interframe distances s(i) makes use of HSV color space using not only an average value but rather a histogram of all three components H, S and V.
  • the method is based on the following steps:
  • the interframe similarity s(i) as defined by Eq. (4) is 1 if two consecutive frames are identical and smaller than 1 if not.
  • FIG. 3 shows a typical plot of the interframe similarity s(i) for s series of some 220 frames.
  • interframe similarity s(i) designates any parameter that expresses the similarity or dissimilarity of consecutive frames.
  • interframe similarities s(i) can be used, as mentioned, in the temporal analysis. In particular, they can be used to calculate one or more of the following three parameters:
  • FIG. 4 shows a typical process of video encoding and decoding as well as a video capture device at the output of the video channel.
  • a source video sequence has a frame rate of 25 to 30 fps.
  • a video transport channel offers lower bandwidth than the input source video sequence would require, and it is therefore necessary to reduce a frame rate of the source even before encoding. This downsampling procedure decreases a video quality.
  • FIG. 4 shows a video capture device at the receiving end.
  • the video capture device works at 25 to 30 fps (i.e. the same frame rate as source video). Increasing the sampling rate to values higher than the one at the video output will not increase video quality since the so called native frame rate (at the input of the encoder) is still unchanged.
  • the downsampling estimation is based on assumption that the capturing rate of the video capture device is the same as the frame rate of the source video sequence. Alternatively, the capturing rate may be higher than the frame rate of the source video sequence.
  • the present method makes use of a histogram approach as explained in the following.
  • the interframe similarities s(i) are compared to a first threshold value k.
  • this first threshold value k is 1. Any value of s(i) ⁇ 1 indicates that the frame contents have changed between frames i ⁇ 1 and i.
  • comparing s(i) to 1 allows to determine the distance between frame changes. This distance is e.g. 1 if the frame rate is the original frame rate, and it can be larger than 1 in the presence of downsampling.
  • FIGS. 5-8 show example histograms of the values H(i) for different downsampling factors DS.
  • the downsampling factor DS is defined by:
  • DS Source_fps native_fps , ( 5 ) which is a ratio of source and native frame rates, the native framerate being the framerate before encoding and the source framerate being the framerate after encoding (capturing rate) and also of the reference video signal.
  • Typical native frame rates in live networks lie between 2.5 and 25 fps, which means that the down sampling factor DS is between 10 and 1.
  • the downsampling factor DS can be calculated from the following equation:
  • DS i ⁇ ⁇ 1 ⁇ H ⁇ ( i ⁇ ⁇ 1 ) + i ⁇ ⁇ 2 ⁇ H ⁇ ( i ⁇ ⁇ 2 ) H ⁇ ( i ⁇ ⁇ 1 ) + H ⁇ ( i ⁇ ⁇ 2 ) ( 6 ) where indices i1 and i2 correspond to the bins with the two largest bin values H(i1), H(i2).
  • Pausing estimation is used to detect events where a video channel causes information loss. In such a situation where video transmission is broken, a capturing device records always the same frame. In this case we talk about so called irregularly repeated frames which are different from those caused by downsampling effect. Downsampling causes periodical frame repetitions whereas pausing is much more annoying and occurs burstwise. Identification of pausing is very important for no reference models since this kind of degradation is very annoying and its impact is always perceived by the viewers.
  • the accuracy of pausing estimation depends on the result of the downsampling calculation.
  • a pausing estimator has to distinguish the irregularly repeated frames from those that are regularly repeated due to downsampling.
  • FIG. 9 shows another example of an interframe distance histogram (derived in the same way as the histograms of FIGS. 5-8 ) two different types of interframe distances can be distinguished. The two highest bars for distances 2 and 3 can be attributed to downsampling, while the bars at distances 5-9 represent irregular frame repetitions or pausing.
  • the pausing parameter p can be calculated from the sum of the bin values H(i) for i above a given threshold.
  • This threshold should be higher than the downsampling factor DS.
  • a guard distance of one unit between bars caused by downsampling and those caused by pausing should be used (in FIG. 9 , bin number 4 is a guard unit).
  • ⁇ i 2 + ceil ⁇ ( DS ) N ⁇ H ⁇ ( i ) ( 7 ) wherein N designates the number of bins.
  • N designates the number of bins.
  • the sum is normalized by the total number of frames, i.e. the pausing parameter p can be calculated from
  • a keyframe or I-Frame is a single frame in a video sequence that is compressed without making reference to any previous or subsequent frame in the sequence, i.e. each keyframe carries full information for creating a frame without reference to another frame.
  • non-keyframes carry incremental information for creating a frame from a previous frame.
  • an I-frame is sent typically every half second in order to enable zapping. I-frames are the only frames in a video data stream that can be decoded by their own (i.e., without needing any other frames as reference).
  • the perceived quality of keyframes strongly differs from that of other frames.
  • the short raises of quality are visible as spikes in the interframe similarity s(i). The larger these spikes are in respect to the remaining signal, the lower video quality of the signal is.
  • the present method uses this property to derive another quality parameter, the keyframe distance parameter KFD from the value of the interframe similarity s(i) at the keyframes.
  • the interframe similarity s(i) presented in FIG. 3 is a good example of this, the periodically arising spikes being indicators for key frames.
  • the keyframes can be identified by checking if the interframe similarity s(i) or a value derived from the interframe similarity s(i) lies in a given range.
  • the second derivative of the histogram similarity signal is taken in an advantageous embodiment of the invention. Calculating the second derivate increases the difference between the spikes and the remaining signal (i.e. works as high pass filter).
  • FIGS. 10 and 11 are two examples for the second derivative of the interframe similarity s(i).
  • FIG. 10 corresponds to a sequence with less compression (higher image quality). Therefore the amplitude of the spikes is lower then of those of FIG. 11 .
  • Keyframes are detected by comparing the second derivative of s(i) (or any other suitable value derived from s(i)) with a lower threshold (lower dotted line in FIGS. 10 , 11 ). If the value exceeds this threshold, the frame may be a keyframe.
  • the lower threshold has been introduced to reject noise produced by normal content changes between successive frames. It may e.g. have a value of 0.03.
  • both figures show one dominating, outlying spike, which indicates a shot boundary (scene change).
  • an upper threshold e.g. 0.3 has been introduced. If a spike exceeds this threshold, it is considered a “false alarm” and the spike is not used in further calculations.
  • Another threshold (lower dash-dotted line) has been introduced to reject noise produced by normal content changes between successive frames.
  • the keyframe parameter (or key frame distance metric) KFD is then calculated by
  • KFD 1 M ⁇ ⁇ s i - TH L ( 9 ) with the sum running over all detected keyframes I, M being the total number of keyframes and TH L the lower threshold.
  • the keyframe parameter KFD of Eq. (9) is the average (mean) of the interframe similarities s(i) inside the band limited by the two thresholds, minus the value of the lower threshold (to align the lowest possible result towards zero).
  • a simplified algorithm for calculating the keyframe parameter may e.g. look as follows:
  • the downsampling factor DS, the pausing parameter p, and the keyframe distance parameter KFD can be derived.
  • Additional parameters S 1 . . . S p describing the spatial quality of the frames can be derived by means of the spatial analysis, as mentioned above.
  • MOS o F ( DS,p ,KFD, S 1 , . . . S p ), (10) where F is a linear or non-linear function in its parameters DS, p, KFD, and S 1 through S p .
  • MOS o M ( p ,KFD, S 1 , . . . S P )+ ⁇ MOS, (11)
  • FIG. 12 shows the relation between the downsampling factor DS and the perceived video quality of a number of subjects. Each connected line represents a source played at different frame rates.
  • results are grouped per downsampling factor and then averaged. These averages are vertically shifted in such a way that a downsampling factor 1 (no downsampling) refers to no difference on the MOS scale.
  • the coefficients k 0 , k p , k KFD , k s1 etc. can be derived by carrying out the following steps:
  • a further improvement can be achieved by taking into account that, if one disturbance becomes strong, the watcher user tends to weigh it in non-linear fasion. For example, in the presence of significant pausing, the average user tends to ignore most other errors in an image.
  • different functions F (in Eq. (10)) or different coefficients M (in Eq. (13)) if a given quality parameter becomes dominant This can be implemented by the following steps:
  • Suitable functions F can again be derived from testing a series of video sequences with a group of users and finding an optimum match.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
US12/094,084 2005-12-23 2005-12-23 Non-intrusive determination of an objective mean opinion score of a video sequence Expired - Fee Related US8212939B2 (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CH2005/000771 WO2007071076A1 (fr) 2005-12-23 2005-12-23 Determination sans ingerence d'une note d'opinion moyenne objective d'une sequence video

Publications (2)

Publication Number Publication Date
US20080284853A1 US20080284853A1 (en) 2008-11-20
US8212939B2 true US8212939B2 (en) 2012-07-03

Family

ID=36946983

Family Applications (1)

Application Number Title Priority Date Filing Date
US12/094,084 Expired - Fee Related US8212939B2 (en) 2005-12-23 2005-12-23 Non-intrusive determination of an objective mean opinion score of a video sequence

Country Status (3)

Country Link
US (1) US8212939B2 (fr)
EP (1) EP1964408A1 (fr)
WO (1) WO2007071076A1 (fr)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8543682B2 (en) 2007-05-02 2013-09-24 Spirent Communications, Inc. Quality of experience indicator for network diagnosis
DE602007007587D1 (de) * 2007-05-25 2010-08-19 Psytechnics Ltd Videoqualitätsbestimmung
DE102007060004B4 (de) * 2007-12-13 2009-09-03 Siemens Ag Verfahren und Vorrichtung zum Bestimmen einer Bildqualität
EP2112835A1 (fr) 2008-04-24 2009-10-28 Psytechnics Ltd Procédé et appareil pour la génération d'un paramètre de qualité vidéo
US8189052B1 (en) 2008-07-29 2012-05-29 Sprint Communications Company L.P. Method and system for evaluating video processing technology
TWI411968B (zh) * 2009-12-31 2013-10-11 Via Tech Inc 圖像特徵計算方法以及圖像檢索方法
ES2511044T3 (es) * 2010-07-30 2014-10-22 Deutsche Telekom Ag Evaluación de la calidad de una señal de video durante la codificación o la compresión de la señal de video
US9143776B2 (en) * 2012-05-07 2015-09-22 Futurewei Technologies, Inc. No-reference video/image quality measurement with compressed domain features
CN103747238B (zh) * 2013-02-20 2015-07-08 华为技术有限公司 视频静止失真程度评估方法和装置
US10521472B2 (en) * 2015-02-27 2019-12-31 Realnetworks, Inc. Composing media stories method and system
US10834406B2 (en) * 2016-12-12 2020-11-10 Netflix, Inc. Device-consistent techniques for predicting absolute perceptual video quality

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5553014A (en) * 1994-10-31 1996-09-03 Lucent Technologies Inc. Adaptive finite impulse response filtering method and apparatus
US5801765A (en) 1995-11-01 1998-09-01 Matsushita Electric Industrial Co., Ltd. Scene-change detection method that distinguishes between gradual and sudden scene changes
US6408024B1 (en) 1999-05-12 2002-06-18 Matsushita Electric Industrial Co., Ltd. Telecine video signal detecting device
US20020186768A1 (en) * 2001-05-14 2002-12-12 Koninklijke Philips Electronics N.V. Video content detection method and system leveraging data-compression constructs
US6496221B1 (en) * 1998-11-02 2002-12-17 The United States Of America As Represented By The Secretary Of Commerce In-service video quality measurement system utilizing an arbitrary bandwidth ancillary data channel
US20030046384A1 (en) * 2001-09-05 2003-03-06 Sudheer Sirivara Server-side measurement of client-perceived quality of service
US6600835B1 (en) * 1999-02-10 2003-07-29 Nec Corporation Moving-picture compressing technique
US20040008892A1 (en) * 2002-07-12 2004-01-15 Mtl Systems, Inc. Apparatus and method for characterizing digital images
US20050094901A1 (en) * 1999-12-06 2005-05-05 Hyundai Curitel, Inc. Method and apparatus for searching, browsing and summarizing moving image data using fidelity for tree-structured moving image hierarchy
US20060072832A1 (en) * 2004-09-28 2006-04-06 Nemiroff Robert S Method and apparatus to detect anchor frames from digital video streams
US7110454B1 (en) * 1999-12-21 2006-09-19 Siemens Corporate Research, Inc. Integrated method for scene change detection

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5553014A (en) * 1994-10-31 1996-09-03 Lucent Technologies Inc. Adaptive finite impulse response filtering method and apparatus
US5801765A (en) 1995-11-01 1998-09-01 Matsushita Electric Industrial Co., Ltd. Scene-change detection method that distinguishes between gradual and sudden scene changes
US6496221B1 (en) * 1998-11-02 2002-12-17 The United States Of America As Represented By The Secretary Of Commerce In-service video quality measurement system utilizing an arbitrary bandwidth ancillary data channel
US6600835B1 (en) * 1999-02-10 2003-07-29 Nec Corporation Moving-picture compressing technique
US6408024B1 (en) 1999-05-12 2002-06-18 Matsushita Electric Industrial Co., Ltd. Telecine video signal detecting device
US20050094901A1 (en) * 1999-12-06 2005-05-05 Hyundai Curitel, Inc. Method and apparatus for searching, browsing and summarizing moving image data using fidelity for tree-structured moving image hierarchy
US7110454B1 (en) * 1999-12-21 2006-09-19 Siemens Corporate Research, Inc. Integrated method for scene change detection
US20020186768A1 (en) * 2001-05-14 2002-12-12 Koninklijke Philips Electronics N.V. Video content detection method and system leveraging data-compression constructs
US20030046384A1 (en) * 2001-09-05 2003-03-06 Sudheer Sirivara Server-side measurement of client-perceived quality of service
US20040008892A1 (en) * 2002-07-12 2004-01-15 Mtl Systems, Inc. Apparatus and method for characterizing digital images
US20060072832A1 (en) * 2004-09-28 2006-04-06 Nemiroff Robert S Method and apparatus to detect anchor frames from digital video streams

Non-Patent Citations (9)

* Cited by examiner, † Cited by third party
Title
Ahmed M. Eskicioglu, "Quality Management for Monochrome Compressed Images in the Past 25 Years", 2000 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2000), Proceedings, vol. 6, pp. 1907-1910.
International Search Report of PCT/CH2005/000771 dated Sep. 28, 2006.
Philip Palmer, "Family Films," Jul. 2003, Scriptwriter, No. 11, pp. 30-33.
Radim Javurek, "Efficient Models for Objective Video Quality Assessment", Dec. 2004, Radioengineering, vol. 13, No. 4, pp. 48-50.
Sabine Susstrunk et al., "Color Image Quality on the Internet," 2004, IS&T/SPIE Electronic Imaging 2004: Internet Imaging V, Proceedings, vol. 5304, pp. 118-131.
Stephen Wolf, Measuring the End-to-End Performance of Digital Video System, IEEE Transactions on Broadcasting, vol. 43, No. 3, Sep. 1997. *
Yong Wang et al., "Content-Based Utility Function Prediction for Real-Time MPEG-4 Video Transcoding," 2003 International Conference on Image Proceedings (ICIP 2003), Proceedings, vol. 2, pp. 189-192.
Zhou Wang et al., "Blind Measurement of Blocking Artifacts in Images," Sep. 2000, IEEE International Conference in Image Processing, Proceedings, vol. 3, pp. 981-984.
Zhou Wang et al., "Video Quality Assessment Using Structural Distortion Measurement" 2002, IEEE International Conference on Image Processing, Proceedings, vol. 3, pp. 65-68.

Also Published As

Publication number Publication date
US20080284853A1 (en) 2008-11-20
EP1964408A1 (fr) 2008-09-03
WO2007071076A1 (fr) 2007-06-28

Similar Documents

Publication Publication Date Title
US6473459B1 (en) Scene change detector
US8488915B2 (en) Automatic video quality measurement system and method based on spatial-temporal coherence metrics
Pinson et al. A new standardized method for objectively measuring video quality
US8737486B2 (en) Objective image quality assessment device of video quality and automatic monitoring device
CN101438594B (zh) 图像分析
EP2866447A2 (fr) Procédé et appareil pour évaluer la qualité d'une séquence vidéo par synchronisation temporelle du flux binaire d'entrée codé d'un décodeur vidéo avec la séquence vidéo traitée obtenue par un décodeur vidéo externe
US8433143B1 (en) Automated detection of video artifacts in an information signal
US8254440B2 (en) Real time scene change detection in video sequences
US8212939B2 (en) Non-intrusive determination of an objective mean opinion score of a video sequence
US8818129B2 (en) Detecting image impairments in an interpolated image
KR20030029909A (ko) 디지털 비디오 품질 측정 방법
EP4111697B1 (fr) Mesure de latence en temps réel de flux vidéo
JP2004529578A (ja) ビデオ信号中の字幕の検出
JP2002262312A (ja) フリーズ・フィールド検出方法及び装置
EP2119248B1 (fr) Concept de détermination d'une mesure de qualité vidéo pour images codées bloquées
WO2010103112A1 (fr) Procédé et appareil de mesure de qualité vidéo sans référence
CA2408435C (fr) Procede pour tester les sequences video
Staelens et al. Viqid: A no-reference bit stream-based visual quality impairment detector
Lee et al. New full-reference visual quality assessment based on human visual perception
Lu Image analysis for video artifact estimation and measurement
Shadiya et al. A Proposed framework for Video Quality Assessment using wavelet and statistical measures
EP2798846B1 (fr) Évaluation de la visibilité de perte de paquets dans une vidéo
Mendi et al. A video quality metric based on frame differencing
Jones COMMITTEE T1 CONTRIBUTION
Mopidevi Video Quality Evaluation using NR metric with Detection of Frame Dips and Drops

Legal Events

Date Code Title Description
AS Assignment

Owner name: SWISSQUAL LICENSE AG, SWITZERLAND

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:JURIC, PERO;WIDMER, RENE;REEL/FRAME:020962/0688;SIGNING DATES FROM 20080319 TO 20080320

Owner name: SWISSQUAL LICENSE AG, SWITZERLAND

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:JURIC, PERO;WIDMER, RENE;SIGNING DATES FROM 20080319 TO 20080320;REEL/FRAME:020962/0688

AS Assignment

Owner name: SWISSQUAL LICENSE AG, SWITZERLAND

Free format text: CHANGE OF ASSIGNEE'S ADDRESS;ASSIGNOR:SWISSQUAL LICENSE AG;REEL/FRAME:022367/0366

Effective date: 20081208

Owner name: SWISSQUAL LICENSE AG,SWITZERLAND

Free format text: CHANGE OF ASSIGNEE'S ADDRESS;ASSIGNOR:SWISSQUAL LICENSE AG;REEL/FRAME:022367/0366

Effective date: 20081208

ZAAA Notice of allowance and fees due

Free format text: ORIGINAL CODE: NOA

ZAAB Notice of allowance mailed

Free format text: ORIGINAL CODE: MN/=.

STCF Information on status: patent grant

Free format text: PATENTED CASE

FEPP Fee payment procedure

Free format text: PAYOR NUMBER ASSIGNED (ORIGINAL EVENT CODE: ASPN); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY

FPAY Fee payment

Year of fee payment: 4

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 8TH YR, SMALL ENTITY (ORIGINAL EVENT CODE: M2552); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY

Year of fee payment: 8

FEPP Fee payment procedure

Free format text: MAINTENANCE FEE REMINDER MAILED (ORIGINAL EVENT CODE: REM.); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY

LAPS Lapse for failure to pay maintenance fees

Free format text: PATENT EXPIRED FOR FAILURE TO PAY MAINTENANCE FEES (ORIGINAL EVENT CODE: EXP.); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY

STCH Information on status: patent discontinuation

Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362

FP Lapsed due to failure to pay maintenance fee

Effective date: 20240703